J . A m . Chem. SOC.1985, 107,1559-1562 to form the dinuclear iron ethylene adduct. (b) Fe2(CzH4)2. Table VI lists the infrared frequencies measured for the diethylene-diiron complex. The spectrum exhibited by this complex is again characteristic of a *-complex. Two isomeric forms of Fe2(C2H4),, denoted as isomer A and isomer B, have been matrix isolated. Isomer A was formed spontaneously upon the cocondensation of iron with ethylene in excess argon. Upon photolysis in the UV the absorptions due to A diminished with the concomitant growth of peaks at lower frequency, suggesting photoisomerization taking place from A to B. It is interesting that the spectra exhibited by Fe2(C2H4),B, and Fe2(C2H4)2rA, are very similar. Furthermore, their carbon-13 and deuterium shifts are extremely close, suggesting very similar structures.
Summary of Results 1. A new type of interaction between a metal atom and an olefin has been demonstrated in this study. Monatomic iron is found to hydrogen bond to ethylene in two distinct forms. One of these forms was shown to be a necessary precursor for the formatior of the insertion product, vinyliron hydride. An oxidative-addi. tionlreductive-eliminationreaction, which is photoreversible, hai been observed: 28C-360 nm
,
Fe(CZH4) ‘ X > 400 nm HFeC2H3
1559
Le., interconversion of ethylene iron to vinyliron hydride and vice versa. 2. The complexes Fe(CzH4)2,Fe2(C2H4),and Fe2(C2H4)2have been isolated in argon and krypton matrices. Only Fe(C2H4)2 was observed in neat ethylene matrices. The interaction between the metal and the olefin in this system follows the DewarChatt-Duncanson model where the ethylene acts as a u-donor through the (C=C) *-orbital and a a-acceptor through the (C=C) a * orbital. Photoexcitation of these p-complexes did not lead to the C-H activation of ethylene. 3. A d-d “forbidden” transition of Fez has been observed for the first time in the infrared region. The activation of this electronic transition is the result of the complexation of the metal dimer to ethylene. A very small deuterium shift, 1 cm-I, has been measured, which is a further proof of its assignment to a Fe,(CzH4) complex.
Acknowledgment. This work has been supported by the National Science Foundation under Grant no. CHE-8315959. We thank Dr. Leif Fredin for the many useful discussions that were carried out during the course of this study. Registry No. 4, 98800-34-1;4 (I3CC-labeled), 98800-35-2;4-d4, 98800-36-3; 9, 98778-48-4;Fe2(C2H4)2 (isomer A), 98778-47-3;Fe2(C2H4)2(isomer B), 98778-49-5;Fe(C,H&, 98778-46-2;”C2H4, 51915-19-6;C2D4,683-73-8; Fe, 7439-89-6;ethylene, 74-85-1.
Matrix-Isolation Studies of the Reactions of Iron Atoms and Iron Dimers with Acetylene in an Argon Matrix. The FTIR Spectrum of Ethynyliron Hydride Ellen S. Kline, Zakya H. Kafafi,* Robert H. Hauge, and John L. Margrave Contribution from the Department of Chemistry and Rice Quantum Institute, Rice University, Houston, Texas 77251. Received May 13, 1985
Abstract: The reactions of acetylene with iron vapors in argon matrices at 15 K have been investigated with Fourier-transform infrared spectroscopy. These studies indicated that at very low concentrations of acetylene and iron the monoiron-acetylene adduct, Fe(C2H2),was formed. The metal is found to be bonded through one of the hydrogen atoms of acetylene. This hydrogen-bonded complex rearranges upon near-ultraviolet photolysis to give ethynyliron hydride, HFeC2H. At higher iron concentrations a diiron adduct, Fe2(C2H,), is formed where the iron molecule is *-coordinated to acetylene. Studies with carbon- 13 acetylene and deuterated acetylene support these findings.
The role of transition metals, particularly iron, in heterogeneous catalytic processes is of great interest. The structure and bonding of small molecules to transition metals have been extensively investigated with use of metal surfaces and metal cluster chemistry.’-I6 In particular, the interactions of acetylene with iron ( I ) Koestner, R. J.; Van Hove, M. A.; Somorjai, G. A. J . Phys. Chem. 1983, 87, 203-213.
(2) Dubois, L.H.; Castner, D. G.; Somorjai, G. A. J . Chem. Phys. 1980, 72, 5234-5240. (3) Baetzold, R. C. Surf. Sci. 1980, 95, 286-298. (4) Foulias, S.D.; Rawlings, K. J.; Hopkins, B. J. Chem. Phys. Lett. 1979, mal-85.
(5) Kesmcdel, L.L.;Dubois, L. H.; Somorjai, G. A. J . Chem. Phys. 1979, 70,2ia0-21aa. (6) Demuth, J. E.Surf. Sci. 1979, 80, 367-387. (7) Demuth, J. E.Surf. Sci. 1980, 93, 127-144. (8) Gavezzotti, A.; Ortoleva, E.; Simonetta, M. Theor. Chim. Acra 1979, 52,209-218. (9) Muetterties, E.L.;Pretzer, W. R.; Thomas, M. G.; Beier, B. F.; Thorn, D. L.; Day, V. W.; Anderson, A. B. J. Am. Chem. SOC.1978,100,2090-2096.
0002-786318511507-7559$01.50/0
surfaces have been studied with new surface science technique^."-'^ It has been postulated that the bonding patterns for unsaturated hydrocarbons in metal cluster chemistry are relatively similar to - ~ ~mono- and those found on transition-metal s ~ r f a c e s ; ’ ~thus, (10) Thorn, D. L.; Hoffmann, R. Inorg. Chem. 1978, 17, 126-140. (11) Nelson, J. H.; Wheelock, K. S.; Cusachs, L. C.; Jonassen, H. B. J. Am. Chem. SOC.1969, 91, 7005-7008. (12) Tatsumi, K.; Fueno, T.; Nakamura, A.; Otsuka, S. Bull. Chem. SOC. Jpn. 1976, 49, 2170-2177. (13) Wang, Y.;Coppens, P. Inorg. Chem. 1976, 15, 1122-1127. (14) Carty, A.J.; Paik, H. N.; Ng, T. W. J . Orgonomet. Chem. 1974, 74,
279-288. (15) Bailey, W.I., Jr.; Chisholm, M. H.; Cotton, F. A,; Rankel, L. A. J . Am. Chem. SOC.1978, 100, 5764-5773. (16) Wheelock, K. S.;Nelson, J. H.; Cusachs, L. C.; Jonassen, H. B. J . Am. Chem. SOC.1970, 92,5110-5114. (17)Erley, W.;Baro, A. M.; Ibach, H. Surf. Sci. 1982, 120, 273-290. (18) Yoshida, K.; Somorjai, G. A. Surf. Sci. 1978, 75,46-60. (19)Brucker, C.; Rhodin, T. J . Cutal. 1977, 47, 214-231.
0 1985 American Chemical Society
1560 J. Am. Chem. SOC.,Vol. 107, No. 25, 1985
Kline et al.
diiron organometallic clusters with acetylene as a ligand have been Table I. FTIR Frequencies (cm-I) Measured for C2H2. l3C2H2,and in~estigated.~'-~~ C,D, in Solid Argon Matrix-isolation studies afford an additional technique whereby vibrational mode C2H2" "C2H2 C2D2a metal atom-hydrocarbon interactions can be studied. Much work u , , C H or CD s-stretch 3373.7 (R) 2700.5 (R) has been reported on transition-metal vapor-acetylene cryoC=C stretch 1973.8 (R) 1762.4 (R) chemistry, notably with nickel, copper, gold, and ~ i l v e r . ~ " ~ ~ u2, u 3 , C H or CD a-stretch 3288.5 3284.9 2441.9 Matrix isolation electron spin resonance studies of copper report u4, C H or C D s-bend 611.8 (R) 505 (R) a copper(0)-monoacetylene complex as shown in I, a a-coordinated v5, C H or CD a-bend 736.7 735.0 542.6 structure consistent with the Dewar-Chatt-Duncanson scheme. u4 + u5, C H or CD s-bend + 1334.5 1324.1 1043.3 Kasai and co-workers have determined that silver atoms do not a-bend interact with acetylene in rare-gas matrices to make a bona fide "Frequencies for Raman active modes of C2H2(g)and C2D2(g)were complex; yet, when silver and acetylene reacted in the gas phase, obtained from ref 36. a silver-acetylene radical of the vinyl form I1 was generated.24 Ab initio calculations support these finding^.^' Matrix isolation Table 11. FTIR Frequencies (cm-I) Measured for (C2HJ2, electron spin resonance studies of gold gave spectral patterns (13C2H2)2,and (C2D2)2in Solid Argon characteristic of complexes I and 11, indicating that both types approx type of of complexes were produced.25 Ozin and co-workers have genvibrational mode (C2H2)2 (I3C2H2)2 (C2D2)2 erated acetylene complexes of nickel and copper atoms in rare-gas 3265.6 3259.8 2424.3 C H or C D stretch matrices.26 Infrared and UV-vis spectroscopies supported a C=C stretch 1968.5 1760.2 Dewar-Chatt-Duncanson bonding picture for both complexes, 1716.8 with the copper-acetylene adduct having less delocalization of C H or CD bend 757.6 749.0 549.6 charge from the metal to the ligand. Ab initio calculations on s-bend + a-bend 1356.4 1347.2 1067.2 manganese-acetylene adducts also predicted a weak a-bonded Table 111. FTIR Measured Frequencies (cm-I) for Iron/Acetylene Products in Solid Argon Fe/C2H2 M
I The literature thus contains several studies of transition metal-acetylene interactions, yet no matrix isolation studies involving iron, one of the most reactive metals, are reported. Therefore, experiments were initiated in our laboratory to study the chemistry of iron atoms with acetylene in an argon matrix. Our results indicate the formation of a new type of adduct 111, in which the iron atom interacts, or adducts, with acetylene through the hydrogen atom. This is the first evidence for a transition metal hydrogen bonding to an alkyne. Diiron was found to interact M.---H-C=C-H I11
with the a-orbitals of the CEC bond of acetylene. The diironacetylene adduct is proposed to be as pictured in IV.
/ .". ,'
I
IV The photochemistry of the adducts was studied via broad-band irradiation of the matrices with a Hg-discharge lamp. It was found that atomic iron photoinserts into the C-H bond of acetylene to give ethynyliron hydride, V. It was also demonstrated that the hydrogen-bonded acetyleneiron adduct is the necessary precursor for this photoinsertion product. H -Fe-C=C--H
V (20) Skinner, P.; Howard, M. W.; Oxton, I. A,; Kettle, S. F. A,; Powell, D. B.; Sheppard, N. J . Chem. SOC.,Faraday Trans. 2 1981, 77, 1203-1215. (21) Ittel, S. D.; Tolman, C. A,; English, A. D.; Jesson, J. P. J . Am. Chem. SOC.1978, 100, 7577-7585. (22) Cotton, F. A.; Jamerson, J. D.; Stults, B. R. J. Am. Chem. SOC.1976, 98, 1774-1779. (23) Hubel, W. "Organic Synthesis via Metal Carbonyls"; Wender, Pino, P., Eds.; Wiley-Interscience: New York, 1968; Vol. I, pp 273-342. (24) Kasai, P. H.; McLeod, D., Jr.; Watanabe, T. J . Am. Chem. SOC.1980, 102, 179-190. (25) Kasai, P. H. J . A m . Chem. SOC.1983, 105, 6704-6710. (26) Ozin, G. A.; McIntosh, D. F.; Power, W. J.; Messmer, R. P. Inorg. Chem. 1981, 20, 1782-1792. (27) Cohen, D.; Basch, H. J . A m . Chem. SOC.1983, 105, 6980-6982. (28) Swope, W. C.; Schaefer, H. F., 111 Mol. Phys. 1977, 34, 1037-1048.
3270.2 1613.4 1600.6 927.1 863.0 760.1 652.8 538.8 3276.2 1976.4 1974.8 1971.9 1969.0 1966.8 1765.0 1762.6 1755.9 1752.5 1740.2 1738.3 1732.7 1729.8
Fe/I3C2H2 Fe/C2D2 (a) Before Photolysis 3263.7 2428.2 1553.1 911.7 871.3 756.9 649.7 540.0
1521.3 775.7 646.3 565.3
(b) After Photolysis 3276.0 2432.8 1905.6 1862.7 1904.1 1861.0 1901.2 1857.8 1898.3 1855.4 1896.4 1853.7 1765.0 1269.4 1762.6 1267.7 1755.9 1262.4 1752.3 1259.3 1740.2 1255.4 1738.3 1250.4 1732.7 1246.3 1729.8 1244.6
assignmentn Fe(C2H2) Fen(CzH2) Fe2(C2H2) Fe,(C2H2) Fe2(C2H2) Fe2(C2H2) Fe2(C2H2) Fe,(C,H2) HFeC,H ( I ) HFeC2H (I) HFeC2H (I) HFeC2H (11) HFeC2H (111) HFeC2H (11) HFeC2H (I) HFeC2H (I) HFeC2H (11) HFeC2H (11) HFeC2H (111) HFeC2H (111) HFeC2H (111) HFeC2H (111)
"Peaks assigned to HFeC2H were grouped in I, 11, and 111, denoting different configurations. Note that within each group, there is more than one set of frequencies, possibly due to different matrix sites.
Experimental Section A complete description of the matrix-isolation apparatus is to be published.29 The metal was placed inside an alumina crucible enclosed in a tantalum furnace. Iron vapors were obtained by resistively heating the tantalum furnace over the range 1150-1450 OC, as determined by a microoptical pyrometer (Pyrometer Instrument Co.). The concentration of iron (Aesar, 99.98%) was varied from 0 to 20 parts per thousand argon (Matheson, 99.9998%) while the concentration of acetylene (Big Three Industries, A.A., 99.6% min.) was varied from 0 to 10 parts per thousand argon. A quartz crystal microbalance was used to determine the rates of deposition of iron, acetylene, and argon. Acetylene and iron atoms were codeposited in excess argon onto a polished rhodium-plated copper surface. The surface was maintained at 15 K by use of a closed-cycle helium refrigerator. After a 30-min deposition, the matrix block was rotated 180° and an infrared spectrum was measured over the range 4000-500 cm-' with a vacuum IBM IR-98 Fourier-transform infrared spectrometer. (29) Hauge, R. H.; Fredin, L.; Kafafi, 2. H.; Margrave, J. L., submitted for publication.
J . A m . Chem. Soc., Vol. 107, No. 25, 1985 7561
FTIR Spectrum of Ethynyliron Hydride
Table IV. FTIR Measured Frequencies (cm-’) for the Isotopomers of Acetylene-Iron in Solid Argon
approx type of vibrational mode CH or CD stretch
Fe(C2H2) 3270.2
Fe(13C2Hz) Fe(C,D,) 3263.7 2428.2
Table V. FTIR Measured Frequencies (cm-l) for the Isotopomers of the Acetylene-Diiron Complex in Solid Argon aomox tvoe of
C=C stretch CH or CD deformations
1553.1 871.3
1521.3
756.9 649.7
565.3
646.3
supported by a recent study carried out by Kafafi et al.,31who have demonstrated that monoatomic iron interacts with ethylene through one or more of its hydrogens. However, one cannot totally rule out the possibility that Fe may be interacting simultaneously with carbon and hydrogen as depicted in VI.
* & + - - - - - ‘ d w - - ~ A
so-*-
1600.9 863.0 760.1 652.8
8E-+Y
855
775 7
5
0
-
(C”‘)
Figure 1. An iron concentration study. Molar ratio of C2H2:Ar 1.1:lOOO. Molar ratio of Fe:1000Ar: A = 0.00, B = 0.26, C = 0.45, D = 1.00, E = 2.10, F = 4.50, G = 9.80, H = 19.10. The peaks located at 1607.9 and 1623.8 cm-l are due to residual water in the vacuum system.
Visible or ultraviolet photolysis was applied after deposition by exposing the matrices to a medium-pressure 100-W Hg lamp. Cut-off filters (500- and 400-nm)as well as a band-pass filter in the range 360 to 280 nm were used in the photolysis experiments. Carbon-13 enriched acetylene (Prochem Isotopes, 75% isotopic purity)
and deuterated acetylene (MSD Isotopes, 99.0% isotopic purity) were also deposited with iron in excess argon in similar experiments.
Results and Discussion FTIR spectra of C2H2,13C2H2,and C2D2in solid argon at 15 K were obtained for reference purposes, and their measured frequencies are listed in Table I. Acetylene has a tendency to “dimerize”, or, more specifically, two molecules of acetylene tend to align and interact with each other in the argon matrix.30 The FTIR frequencies for these dimers were measured and are recorded in Table 11. The acetylene concentration was kept very low (1 part per thousand) in order to eliminate or minimize the formation of (C2HJ2. When iron was cocondensed with acetylene in an argon matrix, new product peaks appeared and are listed in Table 111. The iron concentration was varied between 0 and 20 parts per thousand in order to determine the nuclearity of the iron-acetylene adducts formed upon the cocondensation of iron with acetylene in argon matrices. This study is depicted in Figure 1. Close inspection of Figure 1 reveals a peak at 3270.2 cm-I observed at low iron dispersion. This peak varied linearly with increasing iron concentration until it stopped growing when the molar ratio of iron to argon reached 1 part per thousand. The peak at 3270.2 cm-I is located at a lower frequency than that of the asymmetric C-H stretching frequency of free acetylene, 3288.5 cm-I, indicative of a perturbed, weakened H-C bond. This frequency has been assigned to the iron-acetylene adduct pictured in 111. No peaks were detected in the 1900-1700 cm-I region, characteristic of an activated and perturbed C=C stretch for a a-complexed metal-acetylene as pictured in 11. Indeed, Ozin reports values of 1729 and 1870 cm-I for the activated mode of the a-complexes M(C2H2) where M = nickel and copper, respectively.26 The absence of any perturbed absorption bands between the v m and the vC4 regions and the observed shift in the C-H stretching frequency suggest that a different type of interaction is taking place between monoatomic iron and acetylene in solid argon. We thus propose that the acetylene-iron complex is a hydrogen-bonded one rather than a a-complex. This is further
-
(30) Pendley, R.D.; Ewing, G. E. J . Chem. Phys. 1983, 78, 3531-3540.
H
,111
\-
M----C=C-H
VI
Iron was also codeposited with carbon-1 3 enriched acetylene and deuterated awtylene to measure isotopic shifts and to determine frequency mode assignments. The measured frequencies for Fe(C2H2),Fe(I3C2H2),and Fe(C2D2)are listed in Table IV. for Fe(C2H2) The absence of absorptions due to the IR active -v may possibly be attributed to its being too weak to be observed. As the iron concentration was increased from 1 to 20 parts per thousand, peaks appeared and steadily increased at 1600.9, 863.0, 760.1, and 652.8 cm-’, as indicated in Figure 1. This new set of peaks was assigned to acetylene-diiron. FTIR frequencies measured for Fe2(C2H2),Fe2(I3C2H2),and Fe2(C2D2)are reported in Table V. Another peak also appeared at 1613.4 cm-I and grew with higher iron concentrations. However, this peak did not maintain the same intensity relative to the peak at 1600.9 cm-’ and is probably associated with an iron cluster interacting with the acetylene. The proposed structure for the diiron-acetylene adduct is IV, although one cannot totally eliminate structure VI1 where the acetylene ligand binds parallel to the M-M bond. This di-
\c=c / / MI Mv I1 metallacyclobutene structure is known for few transition-metal complexes such as [Pt2(PhC2Ph)(cod),] 32 (cod = cyclooctadiene) .33 In almost all and [Rh2(~2(q1-CF3C2CF3))(CO)2(qs-C,H,),I dinuclear metal-acetylene complexes, however, the acetylenes are found in a p2-q2 fashion to give an M2C2quasitetrahedral structure with substantial lengthening of the C=C bond?JO The interaction of the carbon orbitals with the metal orbitals can be viewed by orbital symmetry. In the orbital interaction diagram for bonding an acetylene to an M2(C0)6fragment, the acetylene provides two donor orbitals, its a-orbitals of a l and b2 symmetry and two acceptor orbitals, a*, of a2 and bl symmetry. The dimetal acceptor orbitals are b, and 2a,, and the donor orbitals are l a , and b,. This is discussed at length in ref 10. There are several acetylenedimetal clusters with neutral ligands in which the acetylene binds perpendicular to the metal-metal b ~ n d The . ~ Fe2(t-Bu2C2)~ ~ ~ (31) Kafafi, Z. H.; Hauge, R. H.; Margrave, J . L. J . Am. Cbem. Soc., preceding paper in this issue. (32) Boag, N. M.; Green, M.; Howard, J. A. K.; Spencer, J. L.; Stansfield, R. F. D.; Gordon, F.; Stone, A.; Thomas, M. D. 0.;Vicente, J.; Woodward, P. J . Chem. SOC..Chem. Commun. 1977. 930-931. (33) Dickson, R S , IGrsch, H P , Lloyd, D J J Organomel Chem 1975, 101, C48-C50
7562 J. A m . Chem. SOC.,Vol. 107, No. 25, 1985
+-+
I280
t------( 1960
1255 1985
”
Kline et al.
+--1775 7
-
(“1)
A photolysis study. Molar ratios of Fe:C,H,:Ar 0.26:l.l:lOOO: A, no photolysis; B, X 1 4 0 0 nm, 0.5 h; C, 360 nm 2 X 2 280 nm, 1 h; D 360 nm 2 X Z 280 nm, 2 h.
Figure 2.
Table VI. FTIR Measured Frequencies (cm-’) for the Isotopomers of Ethynyliron Hydride in Solid Argon approx type of vibrational mode
HFeC,H
HFe”C,H
DFeC,D
C H or C D stretch C=C stretch
3276.2 1976.4 1974.8 1765.0 1762.6
3276.0 1905.6 1904.1 1765.0 1762.6
2432.8
FeH or FeD stretch
1862.7 1861.0 1269.4 1267.2
2c. Further UV photolysis caused the bleaching of most peaks with the further growth of the set of peaks at 3276.2, 1974.8, and 1762.6 cm-’, respectively. The product peak at 1762.6 cm-’ is in the region characteristic of the Fe-H stretch, as determined by comparison with known values for matrix-isolated HFeOH,34 HFeCH3,35and HFeC2H3.31A large deuterium shift was measured for this frequency in support of its assignment to vFeH. The product peak at 3276.2 cm-I is in the C-H stretching region with a deuterium isotopic shift of 843.4 cm-I. The product peak at 1974.8 cm-’ is in the uc4 region with a carbon-13 isotopic shift of 70.7 cm-]. These FTIR frequencies are summarized in Table VI. The above results indicate that this photoproduct contains an Fe-H, a C=C, and a C-H bond. We propose that photolysis has caused the insertion of atomic iron into the C-H bond of acetylene forming ethynyliron hydride of structure V. This result is analogous to the photolytic study carried out on the monoiron-ethylene adduct Fe(C2H4),in which instance iron photoinserted into one of the C-H bonds to give ethenyliron hydride, HFeC2H3.3’The Fe-C stretch of HFeC2H was not observed in the present study. It is plausible that vFeC for V is beyond the range of the mid-IR region, 4000-500 cm-l, investigated in this study. Kafafi et report a value of 507.2 cm-’ for vFeC in HFeC2H,. No photoproduct peaks were observed for the Fe2(C2H2)adduct in spite of its depletion upon photolysis, which suggests that acetylene-diiron may have just photodissociated. The additional bands in the C=C and Fe-H stretching regions observed in the initial near-UV photolysis must be due to the same photoinsertion product, since further photolysis caused their disappearance with the further growth of the HFeC2H peaks. A possible explanation for this spectrum is that there is more than one geometrical isomer of HFeC2H. Thus, HFeC2H may be trapped or isolated in different configurations in different matrix sites where prolonged UV photolysis causes its rearrangement to the most favored geometry, which is expected to be the linear structure. Structure V has direct analogues in organometallic clusters, most notably, that of the iron [bis(dimethyIphosphino)ethane]hydridoacetylide complex,2’ VIII. VI11 has an infrared spectrum (hexene) with uFeH of 1725 cm-l and C+u of 1894 cm-]. In this
(CO)6 complex has the acetylene ligand roughly perpendicular to the metal-metal bond, with a slight twisting of 4-5O.I0 A double bond exists between the two iron atoms.22 These results all support the structure of a bridging acetylene perpendicular to the metal-metal axis. Most dinuclear metal-acet ylene complexes have carbon-arbon H I bond lengths between 1.31 and 1.38 A, characteristic of a double bond. Further support for acetylenic ligands with double-bond character in these dinuclear metal complexes is found in their infrared spectra. For example, [Ni2(~2(q2-Me3SiC2SiMe3))(cod)2] has a frequency of 1526 cm-I, and [Ni2(p2(q2-CF3C2CF3))(tB u N C ) ~ has ] a frequency of 1562 cm-l, characteristic of a carC bon-carbon double bond.32 The 1600.9-cm-’ peak showed a I carbon-13 shift of 46.9 cm-’ and was assigned to the C = C stretch H of the matrix-isolated Fe2(C2H2)adduct. The lower peaks are VI11 associated with C-H deformation modes. The C-H stretch of configuration the HFeC2H portion of the complex is colinear where the dinuclear adduct was not observed. both the hydrogen and the ethynyl group occupy two of the vertices The spectra in Figure 2 summarize the results of the photolysis of the octahedral structure. By analogy with this transition-metal study carried out on the Fe/C2H2system in solid argon. Photolysis complex we propose that the most stable form of ethynyliron with a 500-nm cut-off filter had no observable effect. Photolysis hydride will have a linear structure with a very low HFeC with a 400-nm cut-off filter produced the monoiron-water insertion bending-force constant. product, HFeOH, characterized by the peak at 1731.7 cm-’ due to the Fe-H stretch.34 This is the result of the photochemistry Acknowledgment. The authors thank the Robert A. Welch taking place between iron and residual water present in the vacuum Foundation and the National Science Foundation for financial system. However, upon photolysis with near-ultraviolet light the support. monoiron-acetylene adduct peak at 3270.2 cm-’ decreased as the photoproduct peaks appeared in different regions of the FTIR (34) Kauffman, J. W.; Hauge, R. H.; Margrave, J. L. J . Phys. Chem. spectrum. These new peaks were concentrated in mainly two 1985.89, 3541-3547. regions: the iron-hydrogen and the C=C stretching regions. (35) Billups, W. E.; Konarski, M. M.; Hauge, R. H.; Margrave, J. L. J . Deuterium and carbon-13 enriched studies of acetylene with iron A m . Chem. SOC.1980, 102, 7393-7394. confirmed the above. A list of the photoproduct frequencies is (36) Shimanouchi, T.; “Tables of Molecular Vibrational Frequencies”; 1971; Vol. 1, pp 71-72. given in Table IIIb, and its FTIR spectrum is depicted in Figure
